Research

Corporate Cash, Investment, and Uncertainty

job market paper

Using firm-level data from Compustat and FISD, TRACE, I empirically discovered that the correlation between firm investment levels and cash flow diminishes when the aggregate uncertainty level is high. Conversely, the correlation between firm cash dividend distributions and cash flow increases under these conditions. This finding is somewhat counterintuitive. I built a simple DSGE model with a flight-to-safety feature to match the empirical results. This occurs because the shareholder shifts her asset holding from a risky asset to a riskless asset, thereby squeezing the firm’s resources and behaving myopic. I also implemented a welfare analysis, and it shows that adding a flight-to-safety feature creates a welfare cost and is thus suboptimal.

You can find the current version of my job market paper here.

Identifying Demand and Supply Shocks

with Dr. Marco Brianti

We used a new identification strategy to identify the effects of aggregate demand and aggregate supply shocks on real U.S. data. The simulation study and the robust test show that this identification strategy works well: (1) the estimated implied impulse response functions are consistent with the model-implied impulse response functions; (2) the estimated shocks and actual shocks are very highly correlated. The empirical results indicate that GDP is less persistent than inflation in response to demand shocks, whereas GDP is more persistent than inflation in response to supply shocks. Also, demand shocks are a more vital driver of boom-bust cycles than supply shocks.

Identifying Oil Uncertainty Shocks

in progress

Crude oil is one of the most crucial production inputs, and oil market turbulence should be a robust driver of the business cycle. Since the oil price is endogenously determined in the economy, identifying shocks associated with oil prices is not easy. This paper proposes a novel focus: oil price uncertainty shock, defined as the second-moment news shock to oil prices. The impulse response analysis and variance decomposition show that the oil uncertainty shock is quantitatively important.